I think it was an aliasing technique to remove some degrees of freedom and therefore prevent the model from becoming over parameterised? Am I close? (no cigar though right?).
I'm with you on this one. I don't know why he put those terms in the model in the first place. Maybe just to make a point that redundant parameters will get set to NA by R? If you try to fit a model with the terms that were set to NA it won't be full rank - doesn't mean you can't do it - just means that the parameters themselves are non-estimable and you could only estimate certain functions of the parameters.
But yeah... why he does it in the first place ... who knows...
I know the thread is closed, but it should probably be closed without leading questions The Ecologist
Not quite sure which part of this thread Dason and I abused, but ok. I'll go along with it!
Edit: If wearing your draws on the outside of a pair of skin tight lycra is an abuse of power, then yes superman does. Now on that lovely mental image, I am off to read chapter 11.
Don't worry. I don't mind that much, we should just start the discussions earlier.
As for the "abuse" issue the thread closing is the closest thing we have to enforce reading deadlines. This way people (like me) actually read the damned chapters
You know what? Lets compromise, I can start an "orphaned chapter discussions thread" where anyone can continue discussions on a closed thread. How does that sound?